The objective of this project is to develop simulation and flow phantom capabilities with which the signal properties of magnetic resonance angiography (MRA) studies of diseased vessels can be understood and, using that information for in vivo studies, define methods which provide improved evaluation of carotid artery disease. This project will develop a comprehensive simulation package to describe image appearance for MR studies of flow through vessels. The simulation will use tortuous and stenotic vascular contours appropriate for diseased vessels. Flow pulsatility, as evaluated from in vivo studies will be used as input to hemodynamic simulations which will generate the trajectories of blood particles moving through 3-D vessels. Magnetization evolution for particles moving along those streamlines will be followed for any specified pulse sequence and the resulting images will be calculated.High resolution MR images will be acquired of carotid plaque specimens removed at endarterectomy. The imaging of these specimens will serve several purposes: It will provide a gold standard by which to compare the relative abilities of in vivo MR angiography, Doppler ultrasound, and conventional x-ray angiography to assess carotid artery disease. The lumenal contours of the flow channel will be measured with high resolutions and will be available as input to the hemodynamic simulations. The sensitivity and specificity of different MRA image acquisition sequences will be evaluated in phantom studies with highly realistic geometries and flow conditions. Phantom studies will incorporate the actual plaque specimens removed at endarterectomy into compliant flow tubes representing the common, internal and external carotid arteries. Effects of patient motion will be eliminated and the sensitivity of images of extrinsic factors such as sequence parameters, and intrinsic factors such as stenosis contours and flow pulsatility will be determined. Commonalities and differences between these measurements and flow simulations will be investigated and exploited to improve image appearance. The imaging protocol determined by simulation and phantom studies to be most suitable will be evaluated in an in vivo trial. A deeper understanding of relevant MR imaging parameters will be provided by the flow simulations which will be easily extended to vascular territories other than the carotid bifurcation. This project is anticipated to result in improved methods of MR angiography, and will assess the ability of MRI to evaluate the flow-lumen and vessel wall of diseased arteries. The role of MRA in staging carotid artery surgery will be evaluated.